Supermicro® Highlights Expanding Kepler-Based GPU Solutions at NVIDIA GTC Japan
Supermicro’s Enterprise-Class GPU Platforms Accelerate Compute Intensive Applications with New NVIDIA Tesla Kepler GPUs
Tokyo, Japan, July 26, 2012 - Super Micro Computer, Inc. (NASDAQ: SMCI), a global leader in high-performance, high-efficiency server technology and green computing, offers the industry’s largest lineup of GPU accelerated servers and workstations based on NVIDIA®Tesla®GPUs. At the NVIDIA®GPU Technology Conference (GTC) 2012 in Tokyo, Supermicro showcases systems supporting NVIDIA’s new Kepler architecture. Supermicro offers the widest variety of hybrid CPU/GPU server solutions in 1U/2U/3U SuperServer®, 4U Tower/SuperWorkstation and 7U GPU SuperBlade® configurations that focus on high-performance, energy efficiency, power conservation and enterprise-class reliability. With a growing line of systems optimized for the new NVIDIA Tesla Kepler (K10) GPU accelerators, Supermicro provides the world’s fastest, most advanced computing platforms for CAD and EDA, computational finance, oil and gas seismic processing, structural mechanics, fluid and molecular dynamics, quantum chemistry and physics, defense and Intelligence analytics, numerical analysis, weather and climate forecasting, material science, media and entertainment applications.
“We are excited to debut our new Kepler-based GPU computing solutions at GTC in Japan where performance and application acceleration with best energy efficiency is top of mind,” said Charles Liang, President and CEO of Supermicro. “Combining our advancements in server design, thermal management and energy-efficient power management technologies with NVIDIA’s Tesla K10 and upcoming K20 GPUs, we are delivering maximum compute performance with superior power efficiency, first to market. With new platforms such as our PUE optimized FatTwin ready to support high capacity hybrid CPU/GPU computing, Supermicro offers an ever expanding range of performance oriented, power efficient server and workstation solutions for the scientific research, engineering and design industries.”
At GTC, Supermicro displays its widely deployed 2U (SYS-2027GR-TRF) SuperServer® featuring an optimized cooling design that supports up to six double-width GPUs, dual Intel® Xeon® E5-2600 processors, up to 256GB, 1600MHz memory and high-speed interconnectivity through onboard 10GbE or optional QDR/FDR InfiniBand. This high efficiency, clusterable HPC solution is the ideal platform for advanced scientific and research superclusters and its venerable predecessor, the SYS-2026GR-TRF, the brains behind Virginia Tech’s HokieSpeed supercomputer, symbolizes this by holding honored positions at 29th on the Green500 and 147th on the Top500. High-end computing at the desktop for scientific and engineering applications is addressed by the 4U (SYS-7047GR-TPRF) SuperWorkstation running an N-Body 2 gravitational simulation utilizing the hybrid computing power of four passively cooled K10 GPUs combined with dual Intel® Xeon® E5-2600 processors. In an NVIDIA® Maximus™ certified configuration, this top of the line workstation offers the industry’s highest GPU count in a workstation combining an NVIDIA Quadro® GPU to accelerate design and visualization tasks with four double-width NVIDIA Tesla C2075 GPUs for simultaneous simulation and rendering. This system offers extreme expandability supporting up to 16 DIMMs for 512GB 1600MHz memory, seven PCI-E slots and eight hot-swap 3.5” HDDs. ASYS-7037A-i targeted at creative design and multimedia professionals will demo video and effects programs in a smaller footprint SuperWorkstation supporting dual actively-cooled double-width GPUs, dual Intel® Xeon® E5-2600 processors (up to 150W), up to 512GB, 1600MHz memory in 16 DIMM slots, four internal 3.5” HDDs, dual IEEE1394 Firewire ports and built-in 7.1 HD audio.
Supermicro’s extensive GPU platforms also include the ultimate scalable and high-density 1U four GPU, dual-processor (DP)SYS-1027GR-TQF featuring non-blocking GPU connections, the value-oriented 1U, two K10 GPU, uni-processor (UP) solutionsSYS-1017GR-TF/SYS-5017GR-TF, 3U two GPU DP SYS-6037R-72RFT+ supporting up to 768GB 1600MHz memory in 24 DIMM slots and high compute-density GPU SuperBlade® (SBI-7127RG) supporting 20x GPUs/20x CPUs in 7U. Supermicro’s wide range of GPU computing solutions support Intel® Xeon® E5-2600 series processors and offer high performance I/O and connectivity with PCI-E 3.0, 10GbE and 4x QDR/FDR InfiniBand. In addition, energy efficiency and overall cost savings are maximized with innovative high-efficiency (95%+) digital switching power supplies and battery backup power (BBP™) module technologies which can eliminate traditional UPS systems.
Browse Supermicro’s total line of high performance, high-efficiency GPU server solutions at www.supermicro.com/GPU .